Explosion-induced domino accidents in the chemical industry, such as the 2005 Buncefield and 2019 Xiangshui accidents, can lead to catastrophic losses. Recent studies commonly use probit models (simplified linear regression models) to predict the probability of accident escalation caused by equipment failure due to overpressure conditions but necessitate distinct equations for different equipment types. In order to simplify the number of models and improve their accuracy, this study introduced three machine learning models (random forest model, convolutional neural network model, and deep neural network model), addressing complex nonlinear relationships that conventional regression models may not fully capture. By model training, the DNN model has the highest accuracy (99 %), followed by CNN (94 %) and random RF (95 %). The DNN model was selected as the optimal data-driven model for equipment vulnerability assessment due to their feedforward mechanism's capability to dynamically align parameters with evolving data distributions. The approach developed can not only predict the probability of equipment damage by integrating values related to peak overpressure and equipment type but also effectively address the accuracy validation issues associated with traditional regression models. Besides, this approach can be considered open source model and more explosion data may be used in the future to further improve the model.

Frequent unpredictable earthquake disasters such as the Turkey Earthquake in 2023 pose an increasing threat to oil tank farms since they may trigger major accidents and domino effects, resulting in casualties, economic losses, and environmental pollution. Unpredictable earthquakes are definitely difficult to prevent and thus resilience strategies such as emergency response should be applied to reduce losses. However, little attention has been paid to the quantitative resilience modeling of oil tank farms, resulting in difficulties in decision-making on resilience assessment and management. Therefore, this study proposes a quantitative seismic resilience model of oil storage tanks by using a dynamic agent-based modeling approach. This approach models the storage tank, active fault, and the environment as three independent agents with their attributes and behaviors. The interaction between agents can also be modeled through disaster evolution rules, and the consequences of interactions can be adjusted through adaptation and recovery strategies. The dynamic propagation of earthquake accidents and the evolution of potential domino effects can be quantified from a bottom-up perspective, thereby quantifying the seismic resilience of oil tank farms. A case study is carried out to illustrate the application of the developed model in oil tank farms and to analyze the sensitivity of different model parameters. The results show that the developed model can dynamically characterize the evolution of earthquake-induced domino effects as well as the emergency and restoration processes, supporting the decision-making on the allocation of resilience measures.